TESTING THE CAUSALITIES BETWEEN ECONOMIC POLICY UNCERTAINTY AND THE US STOCK INDICES: APPLICATIONS OF LINEAR AND NONLINEAR APPROACHES

Abstract

This study aims to investigate the causal relationships between the economic policy uncertainty (EPU) index and the US stock indices the S&P 500, Dow Jones and Nasdaq 100 over the period of 1985M10–2016M12. To this aim, it is applied both linear causality and the nonlinear rolling window causality tests. The long-run analyses are also applied to reveal the elasticities of the US stock indices. The empirical findings indicate that both the linear and the nonlinear approaches support the evidence of causal relationships from the US EPU index to all the US stock indices. Additionally, it is also found some evidences of causal relationships from these three US stock indices to the US EPU index especially between 2008–2013 in the latest financial crisis and in the late 2016. The results indicate that the Nasdaq 100 index is mostly affected from the US EPU index both in the short and long-runs.